The neuropsychological effects of type 1 diabetes, its treatment and its metabolic consequences remain controversial. Because intensive diabetes therapy (IDT) greatly increases the risk of severe hypoglycemia, it is widely believed -- though unproven -- that cognitive skills may be compromised in patients treated with IDT. Other research suggests that persistent hyperglycemia may affect the integrity of the central nervous system and thereby compromise cognitive ability. To resolve these issues, we are proposing to study a well-characterized, carefully followed group of patients with type 1 diabetes who were treated with either conventional therapy, or with IDT as participants in the Diabetes Control and Complications Trial (DCCT), and were evaluated in terms of medical and neuropsychological outcomes. These individuals (n = 1409) are currently being followed for an additional 12 years, as part of the Epidemiology of Diabetes Intervention and Complications (EDIC) study. By the end of that study, the cohort will have been followed for 18 years on average and will be approximately 46 years old with an average duration of diabetes of 24 years. Biomedical measures and assessments of complications, including retinopathy, neuropathy, and cardiovascular risk factors and outcomes are regularly and routinely assessed in the cohort and will be available for our analyses, as will genetic markers thought to be associated with cognitive functioning e.g., angiotensin converting enzyme (ACE) and apolipoprotein E (APOE) genotypes. We will reassess cognitive ability at the end of the EDIC, using the original DCCT neuropsychoiogical test battery, augmented by measures that will allow us to characterize patients on several aspects of cognitive functioning previously found to be affected by metabolic abnormalities of diabetes. Our research aims to address three primary hypotheses: 1) The group treated with intensive diabetes therapy (IDT) during the DCCT will show less of a decline on six domains of cognitive function relative to DCCT entry compared to those in the conventional therapy group. 2) Individuals with better prior glycemic control, as indexed by a lower mean HbAlc over the duration of the DCCT and the EDIC, will manifest less cognitive decline relative to DCCT entry, particularly on domains of psychomotor efficiency (processing speed) and simple motor speed. 3) Individuals with fewer episodes of severe hypoglycemia over the duration of the DCCT and EDIC will show less of a cognitive decline relative to DCCT entry, particularly on measures of abstract reasoning, block design, object assembly and symbol digit learning. In addition, we will examine the effects of key demographic and biomedical variables such as age of diabetes onset, gender and current age, history of hypertension, lipid levels and cardiovascular status (assessed by carotid intimal-medial thickening and calcification of coronary arteries) on cognitive functioning. The proposed study is in unique position to provide important evidence about the effects of diabetes and its treatment on central nervous system functioning and will extend our understanding of the interplay of biomedical factors and cognition.